China and Global Labor Standards: Making Sense of Factory Certification1 Tim Bartley (
[email protected]), Ohio State University Lu Zhang (
[email protected]), Temple University Forthcoming in China and Global Governance: The Dragon’s Learning Curve, ed. by Scott Kennedy. Routledge, 2013.
The globalization of production has spurred important debates about transnational business and labor rights. Fearing that competitive global production systems would degrade labor conditions, labor rights activists have often called for the globalization of labor standards (Chan and Ross 2003). Attempts in the 1990s to add a “social clause” to the General Agreement on Tariffs and Trade (GATT) and to incorporate labor standards into the World Trade Organization (WTO) failed, leading activists to look for other ways of enforcing labor standards. The traditional international regime for labor rights and standards, based on ILO conventions, is symbolically important but lacking in enforcement power. Yet a different sort of regime has emerged as firms and NGOs have developed private labor standards for global supply chains. In response to anti-sweatshop activism in North America and Europe, large retailers and brands in the apparel, footwear, consumer electronics, and food industries have developed codes of conduct for their suppliers, have sent auditors to assess compliance, and have sometimes encouraged suppliers to get certified as having decent labor conditions. Groups like the Fair Labor Association (FLA), Ethical Trading Initiative (ETI), and the Business Social Compliance Initiative (BSCI) have arisen to oversee factory auditing, and organizations like Social 1
Accountability International (SAI), the Worldwide Responsible Accredited Production (WRAP) program, and the International Council of Toy Industries (ICTI) have developed certification programs. As of mid-2012, over 3,000 facilities worldwide are certified to the SA8000 standard developed by SAI, and approximately 1,500 are certified to WRAP standard developed by the American Apparel and Footwear Association. The notion of third-party certification has proven especially intriguing to scholars, in part because it has become a prominent form of “private regulation” across a number of different industries and issues, from labor to environmental sustainability to fair trade (Bartley 2011; Cashore, Auld and Newsom 2004; Vogel 2008). Certification can in some situations improve firms’ market positions or help them respond to activist pressure. Yet evidence is mounting that even credible, third-party certification programs rarely live up to their image (Seidman 2007) and that factory auditing often fails to generate compliance (Locke, Amengual and Mangla 2009). Serious questions remain about what exactly global standards and certification are and are not capable of doing, their effectiveness in different industries and nations, and their role in broader fields of global governance. China looms large over all of these questions, given its huge role in global production and its growing impact on global governance. Over the past two decades China has experienced explosive economic growth, surging exports, and massive foreign investments, becoming “the world’s factory.” In the process, “made in China” has become synonymous in the minds of many with sweatshop conditions and the repression of labor rights. Yet China has also become the epicenter of attempts to improve factory conditions through voluntary standards, auditing, and certification. The rapid growth of Chinese exports in apparel, footwear, toys, and consumer electronics, especially after China’s entry into the WTO in 2001,drew a great deal of attention to 2
issues like forced labor, appalling working conditions, and harsh militaristic styles of managers in export-oriented factories, and the exploitation and marginalization of migrant workers. By the early 2000s, most large American and European brands and retailers were engaging in some type of factory auditing in China, whether done by their own compliance staff or by external auditors. Some firms were also calling for suppliers to be certified by WRAP or to the more stringent standards of SA8000. Certification expanded rapidly in China, as did the broader discourse of Corporate Social Responsibility (CSR). Domestic industry and government actors soon developed their own program, the China Social Compliance CSC9000T standard.2 In this chapter, we examine the dynamics and possible impacts of factory certification in Chinese export-oriented consumer products industries. We address several questions: How have initiatives originating outside China—most notably, the SA8000 standard—been shaped by the Chinese context and potential competition from domestically-driven standards (like CSC9000T)? What are the circumstances in which factories get certified, and to what extent does being certified reliably indicate compliance with standards? To what extent are there clear differences between certified and non-certified factories? Although there is a great deal of discussion of CSR in China, many of these basic questions have remained unanswered. Our attempt to answer them contributes both to the discussion of private forms of global governance and to the questions about China and global governance that are central to this volume. To answer these questions, we draw on approximately 50 interviews with various players involved with private labor standards in China—auditors, factory managers, compliance staff for international brands, migrant labor NGOs, and workers—conducted in 2010 and 2011, mainly in Guangdong province, Shanghai, and Beijing. In addition, we analyze new data from a survey of managers in manufacturing firms in Guangdong province. In general, we find that factory 3
certification’s role in shaping labor conditions in China is quite circumscribed. In the worst cases, factories have been certified despite falling well below the purported standards. Even in the best cases, where certification does seem to mark above-average factories, there are severe limits on what certification can achieve. In what follows, we first examine the terrain in which certification has operated in China, then look more closely at certified factories, particular challenges in the certification process, and the extent of differences between certified and non-certified factories. We conclude by discussing the links between certification and the evolution of labor rights and regulation in China, as well as the ways in which our domain of labor standards contributes to the broader analysis of China and global governance.
The Contested Terrain of Labor Standards in China It is widely held that with the mobilization of China’s millions of cheap and disciplined workers who have no independent trade unions3 and right to strike, a “race to the bottom” has been unleashed, producing an endless downward spiral in labor standards and labor rights around the world. Clearly, the labor situation in China has posed great challenges for international labor advocates. Yet in recent years, China has witnessed a rising tide of labor unrest with the mass movement of capital into China and the deepening of marketization and commodification of labor. According to official Chinese government figures, “mass incidents” increased rapidly from 8,709 incidents in 1993 to 87,000 incidents in 2005, among which about one third were labor-related (Yu 2007).4 The number of labor disputes also increased dramatically during this period, from 48,121 in 1996 to over 350,182 in 2007 (National Bureau of Statistics and Ministry of Human Resources and Social Security 2010). Faced with mounting labor unrest, the Chinese 4
central government passed three new labor laws in 2007, in an attempt to stabilize labor relations and pacify disgruntled workers. In 2008, one year after the implementation of the new labor laws, 1.2 million workers filed over 693,465 labor dispute cases with Chinese authorities, a 98 percent increase from 2007 (ibid). The new labor laws is said to have raised awareness of rights among Chinese workers (especially among the young generation of migrant workers) who are now more willing to stand up and defend their rights through formal legal system (China Labor Bulletin 2009). Against the backdrop of Beijing’s attempt to change from an export-led development model based on cheap labor to a more balanced one based on domestic consumption (and thereby higher wages), Chinese workers’ rising resistances coupled with their growing bargaining power, derived from a labor shortage and changing demographics, have led to ongoing changes in the balance of power between labor and capital at the workplace.5 Clearly, labor relations in China have become quite contentious. The terrain of international labor standards has at times also been quite contentious. The SA8000 standard played a particularly important and provocative role in introducing labororiented CSR to China. This program was founded in 1997 by the Council on Economic Priorities, a New York-based non-profit organization focused on socially responsible investing and shopping.6 SA8000 quickly became so central to the discourse on CSR in China as to lead Certification and Accreditation Administration of China (CNCA), which is charged with approving certification bodies, to seek to clarify that “SA8000 certification is not the same thing as CSR” (Worker's Daily 2004). There are several reasons why SA8000 made such a splash—symbolically at least—in China. First, it mimicked the style of ISO standards (such as ISO 9000 and ISO 14001) which had become a de facto requirement for many Chinese exporters. Second, the SA8000 standard 5
generated more controversy than the codes of conduct of individual companies or programs like WRAP or the FLA because it sought to address the problem of freedom of association in China. Most other codes of conduct endorsed freedom of association but were silent on the question of how this could be implemented in China or other countries where freedom of association is legally restricted. SA8000 called for “parallel means” of worker representation (such as worker committees) where trade unions could not be independent. While this provision earned SA8000 a modicum of credibility among international labor advocates, it also fed into a reaction from Chinese industry and government officials, who framed SA8000 as an illegitimate foreign intrusion. As one labor scholar put it, “before 2005, the government was very defensive about SA8000 . . . [and] saw it as trade protection measure.”7 The People’s Daily warned that “the voluntary standards could become a trade barrier that consumes the profits of Chinese exporters and denies them their biggest advantage in foreign trade, inexpensive labour” (People’s Daily 2004). In this context, domestic actors developed their own set of voluntary labor standards. At the request of Chinese, Taiwanese, and Hong Kong-based companies, the Chinese National Apparel and Textile Council (CNTAC) introduced its CSC9000T standard in 2005.8 This was a clear attempt to appropriate the discourse of international standards, as well as to tailor them to the realities of Chinese manufacturing. The CSC9000T standard focused on management systems more than outright performance, on domestic labor law more than international norms, and on gradual improvement more than pass-fail compliance. The initiative’s 2006 annual report included an astute analysis of the limitations of existing auditing and certification models, noting the drift toward falsification, the problem of individual auditors’ skill and integrity, and the contradictions between brands’ compliance and sourcing practices.9 Yet CSC9000T also 6
reinforced the role of the ACFTU as a legitimate mode of worker representation and treated the use of formal labor contracts as more of a goal than a requirement (CSC9000T Annual Report 2006). Although the program does not certify compliance, it did include procedures for internal and external evaluation of factories, and by the end of 2007, some auditing firms had been named as evaluators and nine apparel manufacturers had been designated as CSC9000T implementers (CSC9000T Annual Report 2007). While many observers initially viewed CSC9000T as an industry- and state-sponsored threat to international standards initiatives,10 CNTAC very quickly began to collaborate with international buyers, building especially on personal connections between CNTAC and the European brands in the BSCI. In 2008, CNTAC and BSCI agreed to explore “compatibility between two systems and thus lay foundation for further cooperation and mutual recognition” (CSC9000T Annual Report 2009). Though CNTAC and SAI had little communication,11 they became indirectly linked by virtue of their shared ties with BSCI, since BSCI and SAI had already come to an agreement whereby BSCI members would use SAI-accredited auditors. Recently, some observers have begun to suggest a three-step process for factories, starting with the CSC9000T program and moving to BSCI-endorsed auditing and then SA8000 certification.12 Several factors seem to have kept CSC9000T from become a strong domestic competitor to international standards like SA8000. International buyers remained more interested in their own systems than in CSC9000T.13 In addition, state actions that might have privileged domestic over international standards were not forthcoming. The CNCA has not explicitly endorsed SA8000. But importantly, neither has it weighed in against SA8000 (as it has in some other international standards projects). CNCA announced in 2004 that SA8000 certifications would require its approval,14 but it did not explicitly promote the Chinese standard or restrict other labor 7
standards from operating in China.15 Overall, SAI has gotten what participants described as a “yellow light” from CNCA—not outright approval but not interference either.16 Some observers continue to describe SA8000 as “illegal” in China, citing the lack of explicit government endorsement, the vexing issue of worker representation, and sometimes other issues as well. One vocal critic complains that SA8000 is not a truly international organization—just an American group—that invokes vague international norms (e.g., ILO conventions, Universal Declaration of Human Rights), and treats Taiwan as a separate country.17 In slightly more measured terms, one specialist noted that “SA8000 asks you to do things in China that are against the law, which gives factories a reason not to do anything.”18 Although tensions remain between international norms and domestic cultures of production, since approximately 2005, the Chinese government has sought to incorporate CSR as a pillar of the “harmonious society” (Lin 2010). Indeed, the party has promoted CSR standards and reporting for private and state-owned companies alike and has embraced some (though not all) international CSR standards.
Opening the Black Box: Portraits of Certified Factories As of late 2011, there were 410 SA8000 certified facilities, employing over 291,000 people in China. Yet many observers suspect that the growth of certification in China has been largely due to lax auditing, and sometimes, to outright fraud. As one factory owner put it, “I believe among every ten certified factories, nine are fake.”19 By all accounts, Chinese factories can rarely if ever meet the standards for maximum hours of work prescribed by Chinese labor law—no more than 44 hours per week and no more than 36 hours per month of overtime—which is required by both SA8000 and WRAP. In one SA8000 certified factory in Guangzhou, 8
employees reported working for approximately 11.5 hours per day on Monday through Saturday, plus at least eight hours on most Sundays, amounting to as much as 77 hours per week.20 Such factories may have achieved certification either by falsifying records or by having the certification audit done during low production seasons. Several SA8000 certified factories have been found, on later inspection, to have serious problems with workplace safety and child labor.21 Speaking about certification programs in general, a compliance official for an international brand suggested that “some factories are getting certified by just hiring a consultant to get them certified,” and that some brands tell factories, “if you’re certified, we won’t bother you with audits.”22 It is clear that essentially no factories can fully meet the letter of the standards—whether these are the basic standards of ICTI and WRAP or the more challenging standards of SA8000. But it is less clear what does happen in factories that are certified. What leads managers to seek certification, and what kinds of changes do they make in order to get certified? How much variation is there among factories certified to the same set of standards? In this section, we address these issues by examining profiles of some certified factories. Admittedly, these may not represent all certified factories or reveal the full implications of certification, but they do help us begin to open what has heretofore been largely a black box to the research community. Comparing SA800 Certified Factories If the SA8000 standard truly serves as a “high bar” that reliably differentiates the very best factories from the rest, then we should expect certified factories to be fairly similar in their performance, at least on the key criteria for certification. Instead, we find evidence of substantial variation among SA8000 certified factories. In one especially egregious case, a factory in Shandong province, making candles and candle-holders for Tchibo was certified despite what 9
was later revealed to be horrendous health and safety practices. The dark factory, lacking in fresh air and rife with fire risks, had workers making candles by hand over a primitive gasoline bottle stove, and workers were not supplied with protective equipment.23 Even in factories with less dire conditions, it is clear that certification does not necessarily mean real standardization of practices. For instance, although worker committees have been promoted by SA8000, we could find little evidence of sustained, empowered worker committees in any Chinese factories. The one documented case of an active, empowered worker committee (Huang 2008; Center for International Private Enterprise 2008 ) was in a factory that has since lost its orders from the brand that had promoted SA8000as well as its visionary manager who had supported the worker elections. This case led observers to doubt that there could be any semblance of a committee left. In fact, many certified factories lack even a rudimentary committee. Auditors do little to check whether there is a well-functioning committee, instead simply asking factory managers if there’s a worker representative. “Auditors don’t really understand the purpose of the committee,” admitted one certification representative.24 A brief comparison of two garment factories in Guangzhou suggests that there can be substantial variation among SA8000 certified factories. This information is based on interviews with several workers, conducted by Chinese research assistants outside the factories. Though it provides only a partial view of each factory, this information is sufficient to identify at least one key difference—working hours. PK Garment25 produces sport shorts and pants for export for Billabong and several other brands. Given Billabong’s promotion of SA8000 as part of its compliance program, it is likely that the brand directed the factory to get certified. The research assistants found three workers to interview, who reported working 11-12 hours per day, six days a week, with one Sunday off per 10
month and shorter hours on most other Sundays. A 70-plus hour workweek like this is certainly not uncommon in Chinese factories, but it contravenes both Chinese labor law and SA8000’s limit of 60 hours per week (though with exceptions under rare circumstances). The workers reported earning 1500 to 2200 RMB per month at PK. Similar wage rates—between 2000 and 2400 RMB per month—were reported by workers at KC Garment Manufacturing, a Hong Kong-owned factory that produces pants for Nike and Seven Wolves (a Chinese brand). Here, however, employees reported working approximately 11 hours per day, five days per week, and just four hours per day on Saturdays. There is reportedly no work on Sundays. This roughly 59 hour work-week, and especially the lack of work on Sundays, marks KC as unusual in Chinese export-oriented industries. In both factories, workers had at least a vague awareness of SA8000 certification, though in neither case did workers demonstrate knowledge of the content and implications of the standards. The reason for the difference between these two factories is not clear, but we suspect it reflects some combination of factors that have been identified in previous research. First, it is possible that workers at KC are misrepresenting their hours of work, perhaps consistent with coaching by management. As many researchers and journalists have discovered, such coaching is rampant, and workers often go along with the charade in fear of reprisals from management, the loss of orders, or preferences for large amounts of overtime compensation. By some accounts, certification raises the likelihood of lying to auditors (and potentially to researchers as well), since the revocation of an all-or-nothing judgment (like a certificate) could have major consequences for the firm. While we cannot entirely dismiss the possibility that the researchers were fed inaccurate information, despite conducting the interviews outside the workplace, we
11
suspect that there are real differences in working hours across the factories that could be accounted for by other factors. It is clear that some brands impose not only more scrutiny than others on their suppliers, but that some brands are more likely to enter into relationships with suppliers that allow for some small but useful degree of trust, cooperation, and joint problem-solving (Frenkel and Scott 2002; Locke, Amengual and Mangla 2009). In addition, scholars have found that differences in the organization of work can sometimes allow productivity gains that allow for somewhat shorter work hours (Locke et al. 2007). Though Nike has not relinquished the sourcing practices that give rise to cutthroat competition and exploitative labor relations, the company has become known for its attempts to build relationships with key suppliers and to push for innovations that can at least potentially increase productivity. Regardless of the reason for the difference, the comparison importantly suggests that something above and beyond factory certification may be necessary to support decent conditions in labor intensive industries. Whether or not certified factories are better on average than non-certified factories—a topic we return to in a later section—it is clear that certification is not a sufficient condition for compliance.
High-Tech Electronics: Certification as a Competitive Advantage A portrait of another SA8000 certified factory illustrates how some firms may seek to use labor standards certification to solidify a competitive advantage. While most firms appear to get certified in response to a specific demand from a buyer, a different path has been taking by at least a handful of Chinese firms—and may become more prominent in a context of labor shortages and movement up the value chain. The following information is based on the factory visit and interviews with company management by the authors in July 2011.26 12
High-Tech Electronics was founded in 2006 as a Sino-Korean high-tech joint venture and became a wholly Chinese private high-tech company in 2007. The company is currently China’s No. 1 mobile TV terminal resolution provider and manufacturer with the largest market share. Because it had the monopoly in mobile communication terminal technology in China until most recently, the profit margin was high—approximately 15-20 percent as of 2010 and reportedly even higher when there was less competition. High-Tech got SA8000 certification in 2009. According to a top manager, the company aimed high from the very beginning and viewed the process of getting certified as a process of self-improvement and becoming a sector leader. It was also viewed as good publicity for a company supplying to well-known multinationals, although most of its customers were not themselves engaged with SA8000. The certification process took roughly one year and involved cooperation between the HR department and the Production Quality Control department, plus several audits by Bureau Veritas (BV)—an SAI-accredited auditor that the company chose for its cost effectiveness (with audits over three years coming to a total cost of about 60,000-70,000 yuan).27 The company had about 110 production workers when we visited it. Brief observation in the factory shops revealed that production workers were most female workers in their early 20s. Although the repetitive simple assembly work did not require high skills, it certainly required carefulness and flexible fingers—typically seen as the attraction of young women workers. In fact, the HR manager told us explicitly that the company preferred 21-22 year old female workers with some working experience in electronics sector. Most workers were from less developed provinces outside Shanghai, such as Anhui, Sichuan, Jiangsu, Hubei, with a minimum of middle school education. About 40 percent workers had high school or technical vocation 13
school education. Some workers had worked at the company since 2007 when the company first began production. Annual labor turnover was around 20 percent. Beginning in 2010, formal employees could sign three-year labor contracts with the company and they were eligible for unfixed labor contract after two renewals of the three-year labor contracts according to the new Labor Contract Law. But most production workers were agency workers from a labor dispatch agency located in a suburb of Shanghai. These workers signed labor contracts with the labor agency and they were not formal employees of High-Tech. Although agency workers were paid by High-Tech directly, their social security premiums were handled by the labor agency based on the Shanghai suburb standard, which was much lower than the Shanghai city standard. The HR department at High-Tech had four full-time staff (including the department manager). The company union was chaired by one of its Vice Presidents. The current worker representative was a regular production worker from Sichuan who was elected by workers. According to the interviewed managers, the worker representative complained several times about meal quality on behalf of workers. Regular working hours were eight hours from 8am to 5pm, with an hour lunch break and a short break in the morning and in the afternoon respectively. Shops were clean with airconditioners (not on though when we visited in July). Production was organized by assembly lines with relatively low automation level. According to the interviewed managers, the company was still in the process of building new shops and adding new lines which could advance the organization of production. Workers’ average wages were about 2200-2500 yuan per month, which was considered more than local average. According to the HR manager, labor costs accounted for less than 15 percent of the company’s total production costs. There was no difficulty in recruiting workers 14
because of the company’s good working conditions and reputation for treating workers fairly. The managers also told us that many workers wanted more overtimes, but because SA8000 set the limit on overtime, workers were acyually unhappy about having fewer overtimes and thus smaller pay checks. As the Vice President of the company commented, “the SA8000 and other international certification programs should consider China’s own unique conditions.”28 Overall, the information we gathered about High-Tech suggests a factory where workers are receiving somewhat higher wages and working few hours than is typical. It is also a company that proactively rather than reactively embraced certification. On the other hand, it appears that the company’s labor contracting practices conflict with SA8000’s requirement that “The company shall not use labour-only contracting arrangements, consecutive short-term contracts, and/or false apprenticeship schemes to avoid fulfilling its obligations to personnel under applicable laws pertaining to labour and social security legislation and regulations.”
Excellent Umbrella: Certification to Meet Customers’ Demand Labor standards certification is, of course, not limited to SA8000. For factories producing for some American brands and retailers, WRAP certification has become common. For those in the toy industry, certification to the International Council of Toy Industry’s (ICTI) CARE program is quite common. While these standards are in some ways less rigorous than those of SA8000, they may in some circumstances help to support improvement in labor conditions in factories, as suggested by the example of Excellent Umbrella. The information is based on the factory visit and interviews with the factory owner by one of the co-authors in June 2011 in the factory in Dongguan, Guangdong Province. The co-author was also allowed to walk around the factory workshops and talk with workers free from management presence. 15
Excellent Umbrella is a Taiwanese-owned company founded in 1993 and exporting mainly to the US, Europe, South Korea, and Japan. It is a supplier to Wal-Mart, Target, Disney, and several other companies. In addition to being certified by ISO 9001, the factory has been certified to the ICTI CARE standard, which calls for decent conditions and systems for continuous improvement in labor management. It was first certified by the ICTI standard in 2006 and re-certified in 2011. Ms. Wang, the factory owner, portrayed certification as necessary to receive orders from customers like Disney and Wal-Mart. In addition to certification, the factory is subject to scrutiny from Wal-Mart’s auditors and the SGS auditors hired by Disney. Observation in the factory and brief interviews with workers revealed working hours of approximately ten hours per day, six days per week. Wages are based on piece-rates, with a base rate of 1100 yuan per month (the local minimum wage) and an average earning around 2,200 RMB per month. The shops appeared clean and had electric fans on, but there were no air conditioners and the shops felt hot. The level of automation was low, with most work being done by hand. Ms. Wang was frank about her company did some “make-up” for auditors, such as having workers studying and memorize prepared answers to auditors. On the other hand, Ms. Wang mentioned that the process of certification and auditing had some positive impact, especially on workplace safety measures and workers’ living conditions in dormitories. For instance, Excellent added a new fire-alarm and protection system, installed extra fans in the shops, refurbished factory toilets, and reduced the numbers of workers living in each dorm room from twelve to eight in the process of preparing for getting certified and auditing. Ms. Wang considered those requirements were not difficult to meet and were good for the factory as well. Meanwhile, Ms. Wang mentioned that after studying the prepared materials to deal with auditors, 16
workers have become more aware of their rights and would ask for better working conditions and overtime payments. As she summarized about the impact of certification and auditing, “It becomes real if a company fakes a long time.” [“jia jiu le jiu bian zhen le.”] This case exemplifies how relatively small but meaningful changes might come about through certification. It also suggests a process by which ceremonial actions to satisfy auditors could gradually shift the operation of the factory.
Comparing Certified and Uncertified Firms While certified factories routinely fail to meet the letter of the standards to which they are certified, this does not necessarily undermine the meaning of labor standards certification. If certified factories are systematically better than other similar factories, even if neither has perfect conditions, then this might be considered a virtue of certification. Indeed, our interviews have revealed that many researchers and practitioners suspect that, despite serious problems, certified factories are on average better than non-certified factories. One labor standards consultant noted that “companies that get certified at least have awareness.” They are at least more likely to “know that fire extinguishers should be there; others may never think about it.”29 The leader of a migrant worker NGO argued that all factories have a long distance to get to the standard of decency and compliance with Chinese labor law but nevertheless suggested that “comparatively speaking, if a factory had certification, protection of workers would be better.”30 A brand compliance official who criticized certification recognized that “certification puts you in a different level because you’ve gone through it. But we would still want to check it [rather than trusting the certification].”31
17
But is it true that certified factories are better on average than others—and if so, in what ways are they better? This important question has not really been tackled by researchers, largely due to the difficulty of getting information on comparable certified and non-certified factories. In this section, we report results from our analysis of new data from a survey of managers in manufacturing firms in Guangdong, led by scholars at the Guanghua School of Management at Peking University. The research focused on five major manufacturing centers (Shenzhen, Dongguan, Zhongshan, Huizhou, and Foshan) and companies in the industries of apparel, textiles, footwear, electronic products, electrical equipment, paper products and printing, and plastics. The research also included some non-manufacturing firms, but these are excluded from our analyses. The sampling strategy was designed to over-sample SA8000 factories, using the public list of SA8000 certified factories as of June 2010. Other firms were selected through business directories and personal contacts, with an attempt to include relevant types of variation across cities and industries. With the help of the Youth League and students from Shenzhen University, researchers contacted the firms, visited, and distributed surveys to managers of production, HR, marketing, and finance in the sampled firms in 2010. In total, approximately 26 percent of the firms in the sample are SA8000 certified. The certified firms in the sample most commonly produce electronics/appliances (36 percent), footwear/sporting equipment (17 percent), and apparel/accessories (14 percent). Since the sampling technique deliberately over-sampled certified factories, these are not meaningful as an estimate of the prevalence of certification across or within industries (all of which would be far lower). This sample does allow us to perform at least rudimentary analyses of how certified and non-certified firms might differ.32
18
Our analyses consider some ways in which certified factories might provide different sorts of working environments than non-certified factories do. Our ability to measure working conditions with this survey of managers is somewhat limited since many questions on this survey asked for subjective evaluations of the factory’s performance and managers have an incentive to report only the best features of their companies. We focus on two measures, for which we expect managers’ answers to be reliable and valid. First, we consider managers’ attitudes toward Human Resource (HR) management. The survey asked managers to rate the importance of the HR office or department in their company.33 Though this does not measure exactly how HR procedures are used, it does tap into an important aspect of managers’ perceptions. Given SA8000’s emphasis on management systems and formal personnel policies, we would expect certified factories to attach greater importance to HR management. As a second measure, we consider whether the factory has a medical clinic (which 43 percent of the sampled firms do). One might expect certified factories to be more likely to have a clinic for several reasons. First, CSR initiatives like SA8000 can be seen as a way of encouraging firms to take on welfare functions (some of which were once provided by the danwei) that migrant workers lack access to in contemporary China. Second, though SA8000 does not require a health clinic, it does emphasize occupational health and safety structures and stipulates that “In the event of a work related injury the company shall provide first aid and assist the worker in obtaining follow-up medical treatment.” A clinic is not the only way to achieve this, but we might expect certified factories to have a formal, reliable way of addressing health issues. In sum, these two measures allow us to assess whether SA8000 certification is linked with more formalized structures for the employment and care of workers. In both analyses, we control for several other factors that might be expected to shape the outcome—namely, firm size (log number of employees), export orientation, foreign ownership, 19
level profit and cash flow (as rated by the surveyed managers), and industry. We also control for other forms of scrutiny or pressure for standards, namely being visited by a government labor inspector in the prior year and having a majority of the factory’s production going to buyers who demand labor standards through a code of conduct or similar policy.
As shown in Table 1, we find that managers in firms that are SA80000 certified do indeed attach greater importance to the HR office/department than managers in uncertified firms do. Being visited recently by local labor inspectors also increases the importance that managers attach to HR management, which most likely reflects the role of the Labor Contract Law in promoting HR systems.34 On the other hand, factories that simply have most of their production going to clients who demand labor standards (i.e., have codes of conduct) do not attach greater importance than others to HR. Neither are export orientation, ownership, or industry linked to the perceived importance of HR. When it comes to the likelihood of having a medical clinic for workers, we find no evidence of a discernible difference between certified and uncertified factories. Clinics are more likely in factories producing footwear or sports equipment and in factories where most clients demand labor standards. There is a suggestive effect of firm size on the existence of a clinic, but it does not reach conventional levels of statistical significance. Overall, these findings paint a mixed picture of the meaning of SA8000 within factories. This type of certification seems to support at least a discursive emphasis on HR management. Certification does not appear to matter for the existence of a medical clinic, thought other types 20
of CSR standards do appear to be linked to the existence of a clinic. In supplemental models, we have also examined some possible business benefits of certification. Notably, SA8000 is not associated with the perceived ease (by management) of recruiting workers. Neither do SA8000 factories appear to be different from others in terms of their long term partners, stability in orders, or growth in clients or orders. These results—while not definitive, due to data limitations—raise important questions about the direct benefits of certification for both workers and firms. At the very least, they remind us that there is enough variation among certified firms that strong patterns in outcomes are hard to find.
Conclusions: CSR, Global Governance, and the Evolution of Labor Regimes in China For scholars of labor, private regulation, and global governance, the case of China illustrates challenges of enforcing labor rights privately via supply chain standards. The space for independent labor activism is heavily constrained, and the culture of factory management in China has evolved around assumptions of strong managers and docile workers. It is not surprising that the direct effects of factory certification have been quite circumscribed. Certification is not a reliable marker of decent labor conditions, and when production demands clash with the requirements of labor standards, the former nearly always win out. Although private labor standards have failed to transform labor conditions in China, our study does suggest that they have played a role in catalyzing discussion of CSR, in the maturing of HR management, and in the discussion of worker representation and communication with management (through committees, for instance). At times, private labor standards have also been one vehicle through which migrant workers have learned about what they are entitled to. But all of these processes have been far more constrained than often claimed. 21
The weaknesses of certification and related forms of private labor regulation are certainly not unique to China. Factory auditing has often failed to spur compliance or widespread improvement (Locke, Amengual and Mangla 2009; Locke, Qin and Brause 2007), and attempts to certify decent factories often generate false positives (Seidman 2007). On the other hand, there is some evidence of higher rates of non-compliance in audited factories in East Asia than in other regions (Locke, Amengual and Mangla 2009). When it comes to certification, our fieldwork suggests that problems with the quality and integrity of the auditing process, while not uncommon elsewhere, may be especially severe in China. “Audit fraud” in Chinese factories has been widely documented (China Labor Watch 2009), admitted by trade associations (International Council of Toy Industries CARE Foundation 2010), and was discussed openly by many of the people we interviewed. The large number of factories being audited and China’s seemingly unparalleled culture of long working hours in labor-intensive industries (Chan and Wang 2004) may be responsible for the high amount of documented audit fraud in the country. Our fieldwork also indicates that oversight from those with an interest in maintaining the integrity of labor standards has been lacking in China. For instance, Social Accountability International, which developed the SA8000 standard, has worked closely with companies, local NGOs, and government agencies in several Central American countries, but local partners in China have expressed concern that SAI and its accreditation body, Social Accountability Accreditation Services, “have not really been thinking about the challenges here.”35 A new SAI project in China may help to improve quality control in factory certification, though the challenges of transforming Chinese labor conditions through voluntary labor standards remain substantial.
22
For scholars of China’s participation in global governance, the case of labor has some interesting features. While the Chinese government has not been active in engaging with global labor standards systems, it has not been passive when it comes to domestic labor conditions and their implications for control and social stability. Furthermore, historically, Beijing’s approach to labor issues has had an indirect but profound influence on global governance of labor. For instance, attempts by international labor advocates to link labor rights with trade were dealt one partial defeat when China’s Most Favored Nation status with the U.S. was de-linked from human rights concerns in 1994. The subsequent decision by the WTO to avoid involvement in labor rights issues was a second major defeat. Though opposition to a “social clause” in the GATT/WTO came from a number of developing countries, it is notable that the contradiction between internationally defined “core labor rights,” especially freedom of association, and domestic labor practice is especially stark in the case of China. The failure of these and other attempts to get inter-governmental global governance of labor is one of the reasons for the rise of the private labor standards that this chapter has examined (Bartley 2007). Indirectly, the Chinese government’s position on labor has influenced global governance, if only by helping to divert it from one arena to another. Overall, the experience of global labor standards in China should not be dismissed but rather should be viewed as part of a dynamic process. The more important question becomes not “are private standards effective?” but rather “what role do private standards play in dynamic and multi-faceted political economy of labor?” Our research has led us to the view that it is possible for private labor standards to productively connect with other forms of labor advocacy, and that the interplay of international norms and domestic governance can sometimes spur new experiments and coalitions for social change. But more important forces in driving social change 23
are workers’ collective actions (whether organized or unorganized) and government enforcement of labor laws and regulations (often in response to the former). As the 2010 autoworker strikes suggested, disruptive wildcat strikes can result in significant gains for workers (e.g., rising wages, improved working conditions, greater responsiveness of unions to workers on the shopfloor) and amplified social effects that catalyze the transformation of work and industrial relations (Zhang forthcoming). Certainly, as Kennedy’s chapter in this book points out, Beijing’s priority in the reform era has been skewed heavily toward supporting rapid economic growth and those interests who most directly help achieve this goal (i.e., managers/capitalists) over social and environment development such as labor standards. However, it is also important to recognize the centrality of maintaining social stability as the Chinese Communist Party’s fundamental political logic. The Party’s pursuit of economic growth is not the goal but a means through which to strengthen the regime legitimacy and to maintain its monopoly of political power. Indeed, there have been clear signs of the Chinese central government reorientation in its development strategy since the mid2000s, when the accumulated social and environmental ills have raised the alarm of the sustainability of the single-minded unbalanced economic growth model and the regime legitimacy. Especially in the wake of the 2008 global financial crisis, the Chinese government has explicitly stated that the primary task of China’s economic structure adjustment is to shift from an export-led growth model based on cheap labor to a more balanced and sustainable one based on “expanding domestic demand” and thereby higher wages for Chinese people (Xinhua 2010). By the end of 2010, almost every province and municipality in China had increased its monthly minimum wage by an average of 23 percent (China Labour Bulletin 2011). The government policy shift, coupled with China’s changing demographics and growing bargaining 24
power and consciousness of the young generation of Chinese workers, has signaled ongoing major changes in the balance of power between labor and capital at the workplace. While it remains to be seen where and how these developments will evolve, it seems clear that “bottomup pressures from workers and concerns about social unrest from ruling groups” that have propelled improvement in labor standards elsewhere throughout the 20th century,36 are also important forces in play in China’s evolving domestic labor regime and its engagement with global governance of labor standards.
25
Table 1: Regression analyses of Human Resource and health infrastructure on SA8000 certification and other selected variables Importance of HR dept. (OLS) 0.333* (2.03)
Health clinic (logistic regression) -0.545 (-0.87)
Gov. inspected in 2009
0.304* (2.21)
-0.229 (-0.43)
Majority of production under codes of conduct
-0.122 (-0.78)
1.298* (2.08)
Log # employees
0.0916 (1.40)
0.376 (1.39)
Majority exported
0.0751 (0.46)
-0.391 (-0.62)
Foreign owned
0.0991 (0.59)
-0.209 (-0.33)
Profit
0.140 (1.16)
-0.302 (-0.66)
Cash flow
-0.0919 (-0.57)
0.694 (1.11)
Ind: Apparel or accessories
-0.155 (-0.73)
0.993 (1.10)
Ind: Electronics or appliances
0.0164 (0.10)
0.0885 (0.16)
Ind: Footwear/sports equip.
-0.0659 (-0.28)
2.615* (2.13)
SA8000 certified
Constant N Omitted categories: Industry: Other
3.211*** (6.11) 88
-4.154* (-2.03) 89 + p < 0.10, * p < 0.05, ** p < 0.01, *** p < 0.001 2-tailed tests, t-statistics in parentheses
26
References Tim Bartley, “Institutional Emergence in an Era of Globalization: The Rise of Transnational Private Regulation of Labor and Environmental Conditions.” American Journal Of Sociology 113, no.2(2007):297-351. Tim Bartley, "Certification as a Mode of Social Regulation," in Handbook on the Politics of Regulation, ed. David Levi-Faur (Edward Elgar, 2011), 441-52. Benjamin Cashore, Graeme Auld, and Deanna Newsom, Governing Through Markets: Forest Certification and the Emergence of Non-state Authority (New Haven, CT: Yale University Press 2004). Center for International Private Enterprise, and Social Accountability International 2009, "From Words to Action: A Business Case for Implementing Workplace Standards," http://www.cipe.org/publications/papers/pdf/SAI.pdf. Anita Chan, "Challenges and Possibilities for Democratic Grassroots Union Elections in China: A Case Study of Two Factory-Level Elections and Their Aftermath," Labor Studies Journal 34, no.3 (2009):293-317. China Labor Bulletin, “Chinese government says labor disputes doubled in 2008,” 11 May 2009, http://www.china-labour.org.hk/en/node/100461. China Labour Bulletin, “Unity is Strength: The Workers’ Movement in China 2009-2011,” October 2011, http://www.clb.org.hk/en/files/share/File/research_reports/unity_is_strength_web.pdf. China Labor Watch 2009, "Corrupt Audits Damage Worker Rights: A Case Analysis of Corruption in Bureau Veritas Factory Audits," www.chinalaborwatch.org. Stephen J. Frenkel and Duncan Scott, "Compliance, Collaboration, and Codes of Labor Practice: The Adidas Connection," California Management Review 45, no.1(2002):29-49. Global Labor Strategies, “Why China Matters: Labor Rights in the Era of Globalization,” April 2008, http://laborstrategies.blogs.com/global_labor_strategies/files/why_china_matters_gls_report.pdf. Yan Huang, "Labor Solidarity in Contract Manufacturing: the Staff Committee Experiment in Xinda Company as an Example," Chinese Journal of Sociology 28, no.4(2008):20-33. Li-Wen Lin, "Corporate Social Responsibility in China: Window Dressing or Structural Change?" Berkeley Journal of International Law 28, no.1(2010):64-100. Richard Locke, Matthew Amengual, and Akshay Mangla, "Virtue out of Necessity?: Compliance, Commitment and the Improvement of Labor Conditions in Global Supply Chains," Politics & Society 37, no.3(2009):319-51. Richard Locke, Thomas Kochan, Monica Romis, and Fei Qin, "Beyond corporate codes of conduct: Work organization and labour standards at Nike's suppliers," International Labour Review 146, no.1-2 (2007):21-40. National Bureau of Statistics (NBS) and Ministry of Human Resources and Social Security (MOHRSS), China Labour Statistical Yearbook (Beijing: China Statistics Press, 2010) Gay Seidman, Beyond the Boycott: Labor Rights, Human Rights and Transnational Activism (New York, NY: Russell Sage Foundation/ASA Rose Series, 2007). Ngai-Ling Sum and Ngai Pun, “Globalization and Paradoxes of Ethical Transnational Production: Code of Conduct in a Chinese Workplace,” Competition and Change 9, no.2(2005):181-200. David Vogel, "Private Global Business Regulation." Annual Review of Political Science 11(2008):261-82. Yu Jiangrong, “Zhongguo de saoluan shijian yu guanzhi weiji,” [“Riots and Governance Crisis in China,”] speeach the University of California, Berkeley, 30 October 2007 . Xiaomin Yu, "Workplace Democracy in China's Foreign-Funded Enterprises: A Multilevel Case Study of Employee Representation." Economic and Industrial Democracy 29, no.2(2008):274-300. 27
Xinhua, “Qiushi zazhi fabiao Li Keqiang tongzhi wenzhang qiangdiao tiaozheng jingji jiegou dui cujin chixu fazhan juyou guanjianxing zuoyong” [Qiushi Magazine Published Li Keqiang’s Article Emphasizing Adjustment of Economic Structure is Critical to Promote Sustainable Development”], 31 May 2010, http://news.xinhuanet.com/politics/2010-05/31/c_12163253.htm. Lu Zhang, From Detroit to Shanghai? Globalization, Market Reform, and Labor Unrest in the Chinese Automobile Industry (New York, NY: Cambridge University Press, forthcoming). 1
We thank the Research Center on Chinese Politics and Business Initiative on China and Global Governance for partial funding for this project. Some of the research was also funded by the American Sociological Association’s Fund for the Advancement of the Discipline and the Summer Research Award and Grant-in-Aid for Research of Temple University.
2
CSC9000T (China Social Compliance 9000 for Textile & Apparel Industry) is a Social Responsibility Management System based on China’s laws and regulations, international conventions and standards, and China’s particular situations.
3
The sole official union in China--the All China Federation of Trade Unions (ACFTU)--is subordinate to the Party-state and is widely considered to be either unable or unwilling to defend workers’ interests in the market economy.
4
The Chinese government stopped publicizing such numbers in 2005. But various sources have pointed to continuing rapid increase in the number of “mass incidents” throughout China.
5
One good illustration of these changes is the success of the 2010 autoworkers’ strikes in winning employers’ concessions, which quickly produced a ripple effect with a wave of strikes across sectors and regions pushing a rapid trend towards wage increases in China.
6
As the anti-sweatshop movement in the U.S. gained steam, this organization brought together several groups of brands with codes of conduct and eventually joined with some—like Toys ‘R’ Us, Avon, Otto-Versand, and the global certification firm Société Générale de Surveillance (SGS)—to create the SA8000 standard and a system for certifying factories.
7
Interview, Beijing, 9 December 2010. 28
8
Interview with compliance program representative, Beijing, 15 July 2011.
9
Similar arguments have been made by critical scholars of private regulation (Locke, Amengual and Mangla 2009; Sum and Pun 2005).
10
Interview with NGO representative, Hong Kong, 17 May 2007.
11
Interview with program representative, Beijing, 15 July 2011.
12
Interview with certification representative, 8 December 2010
13
Interview with factory managers, 25 November 2010.
14
Pacific Institute 2005, http://inni.pacinst.org/inni/inni_online_update_9.htm
15
Interview with labor standards consultant, Beijing, 21 July 2011.
16
Interview with certification representative, Beijing, 8 December 2010.
17
Interview with CSR researcher, 1 December 2010.
18
Interview with certification official, 5 January 2011.
19
Interview, Dongguan, 24 June 2011.
20
Interviews with workers, Guangzhou, November-December 2010.
21
Interview with certification representative, Shenzhen, 12 November 2010.
22
Interview, Shanghai, 13 July 2011.
23
Interview with certification representative, Shenzhen, 11 December 2010.
24
Interview with certification representative, Shenzhen, 12 Novemner 2010.
25
For this and other factories where interviews were conducted, we are using pseudonyms for the factory names.
26
We were introduced to the company by the local government as an international research team studying the impact of SA8000 in China. The company’s Vice President, the Human Resource (HR) manager, and the Production Quality Control manager met with us and provided the 29
information. The interviews were conducted in Chinese and were recorded and translated into English. We also visited the company workshops accompanied by management. 27
According to China Labor Watch (2009), BV has also been much maligned for the quality of its audits, probably more than any other auditor in China.
28
Interviews, Shanghai, 11 July 2011.
29
Interview, Beijing, 21 July 2011.
30
Interview, Guangzhou, 20 December 2010.
31
Interview, Shanghai, 13 July 2010.
32
Assessing whether certification caused any observed differences would require additional
inquiry and is beyond the scope of this chapter. 33
We take the mean of their responses on a scale from 1 (does not matter) to 5 (very important).
34
Interview with a labor standards consultant, Beijing, 21 July 2011.
35
Interview with certification representative, Shenzhen, 12 November 2010.
36
On the parallel, see Global Labor Strategy (2008): 6
30
